Do investors love their losers more than their winners? In their January 2007 paper entitled “The Effect of Prior Beliefs and Preferences on Information Processing in an Investment Experiment”, Jeremy Ko and Oliver Hansch use a stock picking simulation to measure the bias investors exhibit when processing new information about stocks they have selected. Each iteration of the simulation involves picking one of two similar stocks that will outperform during the coming week. Using results from simulations involving 99 total participants in 2003 and 2004, they conclude that:
- People appear to pay more attention to new information (report news more frequently) when losing a bet than winning. They passively accept good information about stock picks and actively scrutinize bad.
- Perception of new information about the current stock pick is more positively biased when incurring a loss rather than a gain. This bias is apparently due more to distorted interpretation than selective collection.
- Perception biases are apparently not just posturing. People are more likely to bet again on a prior pick when realized returns are negative than positive, and they tend to increase the size of the bet on it.
- Processing of new information about the stock not picked appears to be unbiased, suggesting that participants in the simulation framed their decision as choosing the good stock rather than avoiding the bad one.
In summary, investors appear to focus on underperforming/losing stocks, cut them more slack and put more money into them.
Investors/traders may want to adopt formal, even-handed rules for tracking news on their holdings.
